Implantation of 111In in the Heusler alloys Pd2MnZ (Z=Sn,Sb,Ge,In) following heavy ion nuclear reactions: measurement of magnetic hyperfine field using PAC spectroscopy

APPLIED PHYSICS AND INSTRUMENTATION

Implantation of ¹¹¹In in the Heusler alloys Pd₂MnZ (Z=Sn,Sb,Ge,In) following heavy ion nuclear reactions: measurement of magnetic hyperfine field using PAC spectroscopy

G.A. Cabrera-Pasca^I; M.N. Rao^I; J.R.B. Oliveira^I; M.A. Rizzuto^I; N. Added^I; W.A Seale^I; R.V. Ribas^I; N.H Medina^I; R.N. Saxena^II; A.W. Carbonari^II

^IDepartamento de Física Nuclear, Instituto de Física Universidade de São Paulo, São Paulo, Brazil

^IIInstituto de Pesquisas Energéticas e Nucleares, IPEN, São Paulo, Brazil

ABSTRACT

Ion implantation of the recoil ¹¹¹In nuclei following heavy ion nuclear reactions ¹⁰⁸Pd( ⁷Li,4n)¹¹¹In and ¹⁰⁸Pd( ⁶Li,3n)¹¹¹In has been used to implant ¹¹¹In probes in the Heusler alloys Pd₂MnZ(Z=Sn,Sb,Ge,In). Perturbed Angular Correlation method was used to study the hyperfine magnetic field in these alloys. Direct implantation of ¹¹¹In probe nuclei was used to great advantage in the present case resulting in large implantation efficiency. Only a few hours of irradiation time with moderate beam current of the order of 400-500 nA resulted in sufficient implanted ¹¹¹In activity on the sample for good quality measurements. The hyperfine field was measured at ¹¹¹In probe nuclei substituting Mn and Z sites as a function of temperature. The fraction of ¹¹¹In nuclei occupying Mn atom sites was found to increases with the annealing of sample at higher temperatures.

I. INTRODUCTION

The radioactive probe nuclei, used in the study of hyperfine interactions with Perturbed Angular Correlation (PAC) spectroscopy, are generally produced through nuclear reactions using particle accelerators or nuclear reactors. These radioactive nuclei are then introduced in to the samples to be studied using a variety of chemical and metallurgical processes or through ion implantation. The ion implantation is usually carried out either by using accelerated radioactive ion beams or through radioactive ions recoiling out of the target following heavy ion nuclear reactions.

The ion implantation process is particularly advantageous because it introduces radioactive probes into an already prepared sample, which avoids extensive manipulation of the radioactive material. In the present work direct implantation of ¹¹¹In nuclei in the samples was achieved through nuclear reaction ¹⁰⁸Pd(⁶Li,3n)¹¹¹In or ¹⁰⁸Pd( ⁷Li,4n)¹¹¹In, using 8 UD Pelletron Tandem Accelerator at the Institute of Physics of the University of São Paulo.

Besides reporting a new and efficient way to introduce ¹¹¹In probe nuclei into samples for PAC measurements, we also show that the method of introducing the probe can influence the final site location of the probe and give different results. In order to test the method of implantation we have used Pd-based Heusler alloys as samples and studied the local magnetism by measuring the hyperfine fields with PAC technique.

II. EXPERIMENTAL PROCEDURE

A. Ion Implantation

The technique of recoil-ion-implantation of PAC probes, following heavy-ion nuclear reactions, has been used efficiently to implant the radioactive isotope ¹¹¹In in semiconductor samples [1]. In this method the authors used a thin foil of Rh as target and bombarded it with a ¹²C beam of 69 Mev. The short lived radioactive nuclei ¹¹¹Sn (T_1/2 = 35 min) and ¹¹¹Sb(T_1/2 = 1.3 min), produced in nuclear reactions ¹⁰³Rh(¹²C, p3n)¹¹¹Sn and ¹⁰³Rh(¹²C, 4n)¹¹¹Sb having large recoil energy ( ~ 7 MeV), exit thin Rh foil (2-3 micrometers) and get implanted on substrate kept behind the target. These short-lived nuclei eventually decay to the desired probe nuclei ¹¹¹In (T_1/2 = 2.8 d). The obvious advantages of these reactions are the high natural abundance of ¹²C and ¹⁰³Rh, 98.9 % and 100 %, respectively and relatively high recoil energy of the product nucleus.

The Pelletron Tandem accelerator at the Physics Institute of the University of São Paulo, used in the present work however, can accelerate ¹²C beam to only 56 MeV, which is not sufficient to produce ¹¹¹In in good yield. Given the limitation it was decided to try alternate reactions like ¹⁰⁸Pd( ⁷Li, 4n)¹¹¹In and ¹⁰⁸Pd( ⁶Li, 3n)¹¹¹In, using ⁶Li and ⁷Li beams with maximum available energy of 32 MeV.

It was also decided to use the natural Pd target having ¹⁰⁸Pd with ~ 27 % of abundance, in the preliminary experiments to implant ¹¹¹In in a series of Heusler alloys of the type Pd₂MnZ(Z=Sn,Sb,Ge,In). These alloys have a cubic L2₁ structure and order ferromagnetically with a magnetic moment of about 4.3 m_B localized on Mn. The Heusler alloys Pd₂MnSb(Sn) have been investigated in the past with PAC spectroscopy [2,3]. The radioactive ¹¹¹In probe, introduced in the samples during its preparation by induction melting of component elements, was found to substitute only the Sn and Sb atom sites. On the other hand when ¹¹¹Ag was introduced in Pd₂MnSn sample through thermal diffusion it occupied the Mn site [4].

In the present experiment heavy ion nuclear reactions ¹⁰⁸Pd( ⁷Li, 4n)¹¹¹In and ¹⁰⁸Pd(⁶Li,3n)¹¹¹In, in which Pd₂MnZ(Z=Sn,Sb,Ge,In) Heusler alloys themselves served as the reaction target, was used to implant the recoiling ¹¹¹In nuclei in to the sample. Calculations made with the program PACE for the fusion-evaporation reaction cross sections shown in Fig. 1 indicated that the integral cross-section for the production of ¹¹¹In is considerably larger for ⁶Li beam compared to ⁷Li at all energies above threshold. The ⁶Li beam was therefore chosen for all the experiments.

Since the Heusler alloys chosen for the experiment all contain Pd as one of the component elements, they themselves served as reaction targets. The samples were cut in to small slices of about 5x5 mm² and 1mm thick and mounted in an especial reaction chamber[5] for irradiation with the ⁶Li beam. The average recoil energy of ¹¹¹In ions being too small ( ~ 1.7 MeV) they all stop in the relatively thick ( ~ 1 mm) sample and get directly implanted in the Heusler alloy. This is a great advantage in the present case, as the implantation efficiency tends to be almost 100 %. With a beam current of 400-500 nA only a short irradiation time of the order of 8-10 hours was found sufficient to implant more than 20 µCi of ¹¹¹In in the samples for a good quality PAC measurement.

III. EXPERIMENTAL RESULTS

Since the Heusler alloys Pd₂MnZ(Z=Sn,Sb,Ge,In) used in the present experiment contain, apart from Pd, also other elements such as Mn,Sn,Sb,Ge and In, all of them in their natural isotopic composition, it was realized that nuclear interaction of ⁶Li beam with these nuclei would produce several other radionuclides apart from ¹¹¹In. Depending on the half-lives and gamma rays emitted in their decay these radionuclides could seriously interfere with the PAC measurements.

The low energy gamma ray spectra for some of the Heusler alloys taken with a Ge(HP) detector spectrometer, 10-12 hours after the end of irradiation, are shown in Fig. 2. All the spectra show gamma rays at 171 keV and 245 keV belonging to ¹¹¹In (T_1/2 = 2.8 d) and a gamma ray at 203 keV belonging to ¹⁰⁹In (T_1/2 = 4.2 h) resulting from nuclear reaction with Pd. The principal gamma rays resulting from nuclear reactions with Sn, Ge and In come from ¹²³I (T_1/2 = 13 h), ⁷⁷Br ( T_1/2 = 57 h), ⁷³Se (T_1/2 = 7 h) and ¹¹⁸Sb(T_1/2 = 5 h) as can be seen in Fig. 2.

Since PAC measurements started about 24-30 hours after the end of irradiation, important conclusion is that none of these additional gamma rays interfered with the PAC experiments since it involved gamma-gamma coincidence measurements of the 171- 245 keV gamma cascade in the decay of ¹¹¹In.

TDPAC measurements were carried out at the Hyperfine Interaction Laboratory at IPEN using a spectrometer consisting of four BaF₂ detectors and associated electronic set up generating simultaneously 12 delayed coincidence spectra. Details about the PAC measurements can be found else where [6,7]. The spectrum taken for the “as implanted” sample of Pd₂MnSn at 295K shown in Fig. 3(a) is a typical one for a radiation damaged sample in which the amplitude of the ratio R(t) » A₂₂G₂₂(t), where G₂₂(t) is the perturbation coefficient, shows rapid attenuation. All the irradiated samples were thermally annealed at 400 ºC for 24 hours before starting the PAC measurements to eliminate or substantially reduce the radiation damage effects. The spectrum given in Fig. 3(b) shows an almost complete recovery of the oscillation amplitude after annealing. A slow attenuation still observed in the curve might be the result of a low frequency quadrupole interaction present in the sample due to some disorder or structural defects but does not interfere in the determination of magnetic hyperfine field.

TDPAC spectra for the Heusler alloys Pd₂MnZ(Z=Sn,Sb,Ge,In) taken at temperatures below magnetic transition temperatures are shown in Fig. 4. Detailed analysis of these spectra showed two magnetic interactions in the case of alloys Pd₂MnSn(Sb,Ge) which were assigned to ¹¹¹In probe occupying Mn and Sn(Sb,Ge) sites respectively.

As expected Pd₂MnIn alloy did not show magnetic interaction. This is due to the fact that ¹¹¹In is substituting some of the In atom in this alloy and the crystal structure of the alloy is such that the In atom is in between the two layers of Mn atoms having opposite spins. Due to opposing spins of Mn atoms there is no net transfer of spin density to the probe resulting in zero hyperfine field at In site. After an additional annealing of the sample at 800 ºC for 12 h, PAC measurements showed a unique frequency in this alloy, which was assigned to ¹¹¹In probe nuclei occupying the Mn sites as shown in Fig. 4(d). The migration of radioactive probe from transition element site to Mn site at higher temperature annealing was observed in all the alloys. This was seen from the increase in the fractional occupation of Mn sites relative to transitional element site after higher temperature annealing.

IV. CONCLUSION

Present experiment has demonstrated that for samples where Pd is one of the components, the process of ¹¹¹In implantation using the present nuclear reaction in thick target is quite efficient compared to conventional methods of introducing the probe nuclei in the sample. The main reason for this is the relatively low recoil energy ( ~ 1.7 MeV) imparted to the reaction product. Most of the ¹¹¹In recoils therefore stop in the target(sample)itself. On the other hand for example, conventional ion implanter which produces radioactive ¹¹¹In ion beam for implantation has very small efficiency due to low beam transmission characteristics of these machines (of the order of 0.1%) or less. As a consequence one needs to use very high specific activity of ¹¹¹In in the ion source. In heavy ion nuclear reaction method, high recoil energies are necessary in order to implant the probe nuclei in substrate kept behind the target. About 50-60% of all nuclei produced in target often get implanted in samples.

For the implantation of ¹¹¹In on samples that do not contain Pd the method will require some modifications. The ¹⁰⁸Pd(⁶Li, 3n)¹¹¹In reactions could be produced in a thin foil of Pd (preferably enriched in ¹⁰⁸Pd) and swift ¹¹¹In ions recoiling out of the foil may be stopped in the substrate placed behind the target at a suitable distance and geometry. The reaction chamber for such experiments is under test. Low energy of ¹¹¹In recoils may pose serious problems however, in terms of the efficiency of the process.

V. ACKNOWLEDGMENTS

Partial financial support for this research was provided by the Fundação de Amparo para Pesquisa do Estado de São Paulo (FAPESP). GAC-P thankfully acknowledges the student fellowship granted by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES).

Received on 6 July, 2005

• [1] E. Bezakova Ph.D. thesis, Australian National University, Camberra, Australia, 1998
• [2] J. Schaf, E.R. Fraga, and F.C. Zawislak, J.Mag. Mag. Mat. 8, 297 (1978).
• [3] J. Schaf, A.F. Pasquevich, W.H. Schreiner, C.C.M. Campbell, E.R. Fraga, and F.C. Zawislak, J.Mag. Mag. Mat. 21, 24 (1980).
• [4] A. W. Carbonari and H. Haas, Hyperfine Interactions 133, 71 (2001).
• [5] M. N. Rao, J.R.B. Oliveira, W.A. Seale, M.A. Rizzutto, R.V. Ribas, J.A. Alcântara Nuńez, D. Pereira, N. Added, E.W. Cybulska, N.H. Medina, R.N. Saxena, and A.W. Carbonari, Braz. J. Phys. 33, 291 (2003).
• [6] W. Pendl Jr., R.N. Saxena, A.W. Carbonari, and J. Mestnik-Filho J. Schaft, J. Phys.: Condens. Matter. 8, 11317 (1996).
• [7] R.N. Attili, R.N. Saxena, A.W. Carbonari, J. Mestnik-Filho, M. Uhrmacher, and K.P. Lieb, Phys. Rev. B 58, 2563 (1998).

Publication Dates

• Publication in this collection
  07 Nov 2005
• Date of issue
  Sept 2005

History

• Received
  06 July 2005